I'm trying to design a tunable microstrip combline filter with its initial f0 at 2.45GHz, with my resonator's electrical length at 45 degrees to provide maximum tuning range. I'm having problems with my CST simulation, where I'm trying to find the unloaded Q-factor of a resonator. I don't trust the results I'm getting (a Q0 of over 1000 for a 18um copper trace on a Rogers substrate, which I thought would be below 300) and am wondering if I'm using CST correctly.
The only thing I think can be wrong with the dimensions is that I might have chosen a PCB with too thin or thick a dielectric thickness. What are some common thicknesses used for applications at 2.45GHz?
The below list is the steps that I've followed. Hopefully something erroneous pops up to someone more experienced than I.
I've set up the simulation for Q0 extraction as follows:
My goal is to get Q0 at my desired resonant frequency, but the mode frequencies CST gave me is not at 2.4GHz. I don't think it's a fault with the dimensions, but I could very well be wrong. I suspect I might have made a mistake in how I set up my simulation.
Please advise me on any possible errors I could be making, or perhaps better ways of going about the simulation.
I can't seem to find anything on YouTube detailing the process I'm embarking on, nor do the papers I've found detail exactly how they arrived at their filter's dimensions from simulation, they simply provide their resulting parameters.
Thank you for your time.
Your detailed description of the process provides valuable insights into your problem. Designing a microstrip combine filter at 2.45 GHz and simulating it to extract the unloaded Q-factor (Q0) is indeed a complex task, and minor mistakes can lead to significant deviations from expected results.
Here are some possible factors that might be contributing to the discrepancies in your simulation and ways to address them:
Remember, the simulation of high-frequency devices can be highly sensitive to minor changes in geometry, material properties, and simulation parameters. Rigorous validation, careful setup review, and model iterative refinement are often necessary to achieve accurate results.
Good luck with your design, and I hope some of these suggestions help you achieve the desired results in your simulations!
Joshua Depiver, thank you for the detailed description of where the process might have gone wrong.
Regarding your second point on boolean operations: I know that I can check "electrical connections" with said option, but supposing it fails or the options to check are greyed out, what could be further steps to troubleshoot the connections?
I tried to forego my electrical length of 45 degrees and adjusted the length until the simulation gave me a resonant mode at 2450MHz, but am concerned that this new length's associated electrical length will need to be adjusted again later in order to add the loading capacitors/varactors.
Could you perhaps point me in a direction where I can find reading for how to know which boundary settings to use in CST?
Lastly:
Do you have any advice for simulating the external Q-factor? My plan is to set up the same resonator in its current position, add a 50-Ohm feed line (directly tapped), boolean insert the feedline into the resonator, then setup a waveguide port.
After the geometry is setup, I'll setup the eigenmode solver, consider ports, and do optimization on the feedline's tap-offset with a goal setup for the q-factor at my designed value.
Does the above sound correct?
Hello,
One help file is attached. Could you please provide more detail about the simulation.
Thanks,
How did you implement the resonating capacitances? Depending on earthplane spacing and bandwidth, waveguide modes below cut off can also do funny things.
For coupling factors, etc., Involving two resonators, the resonant frequencies without coupling must be exactly the same if you want to use the standard equation for coupling factor [k=(fo^2-fe^2)/(fo^2+fe^2)].
R.Levy published several excellent must-read articles on comb line filters.
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix